Current State

When using Data Manager to import user objects into Assets, empty cells in the source CSV are handled differently depending on the environment:

• Sandbox site: Empty CSV cells are represented as (Empty) after the cleanse step. During merge into the object class, these values correctly clear existing attribute values on the object.

• Production site: The same empty CSV cells are represented as (null) after the cleanse step. During merge, (null) values do not clear existing attribute values — including attributes linked to an object (object reference attributes) — even when the import settings for Empty values and NULL values are both set to Remove.

This means that despite using identical spreadsheets and identical import configurations, the merge behaves differently across environments, making it impossible to reliably cleanse object attributes in production.

Proposed Future State

• Empty CSV cells should be represented and handled consistently across all environments (sandbox and production).

• The (null) representation in production should behave identically to (Empty) in sandbox — i.e., when NULL values: Remove is configured, object reference attributes (not just text attributes) should also be cleared during merge.

• Alternatively, Data Manager should normalize both (Empty) and (null) to a single internal representation so that import settings apply uniformly regardless of environment.